Stanford researchers used the Evo 2 generative AI model to design thousands of bacteriophage genomes, synthesised nearly 300 candidates, and identified 16 that showed strong E. coli-killing performance in laboratory tests using ΦX174 as the template. Evo 2 generates full phage genomes from short DNA snippets, with a computational framework filtering candidates before costly chemical synthesis and assays. Some AI-designed phages demonstrated higher fitness than native ΦX174, validating whole-genome design capabilities. A 16-phage cocktail rapidly defeated E. coli resistant to native ΦX174, suggesting new options against stubborn infections such as MRSA and Pseudomonas aeruginosa. Evo 2’s open-source release sparks safety debates while enabling broader therapeutic and synthetic biology research.
This update represents a notable development in the Ai sector. Organizations and founders tracking this space should evaluate potential strategic and technical implications on their operations.